The Dog That Never Barks
by David Hawkins
My background and insight
I read natural science at university, many years ago now, and the scientific method is, to my mind, the best way of determining the validity of contentious issues.
After college, I trained and qualified as a Chartered Accountant (CPA in US terms) and used those skills in industry as Financial Controller for the UK subsidiary of a well known food manufacturer. My team kept the books and produced monthly management accounts and the annual accounts for audit. We also spent a lot of time producing forecasts to indicate to management how well the business expected to do in coming months and years.
My point in detailing the responsibilities of my old job here is that the financial results for any particular time period are nearly meaningless by themselves: they acquire context by being compared to other similar results, either the previous period, the same period in the previous year or against forecast. Whilst the previous year is interesting, the most useful comparison is always going to be against the forecast: what you expected this period to do compared to how it has actually turned out.
Many readers will, by now, be ahead of me and can see why I mention this part of financial management here. The similarity between management accounting and climate science is clear: both record and publish actual results and both make forecasts of what will happen in the near or more distant future. The difference, to my eyes, is that management accountants always provide meaningful context to their reporting of actual results and that climate scientists rarely do, and then never provide the most relevant comparisons.
Let me justify that strong statement. The actual temperature for a month or year is often compared by the meteorological authorities (and then reported by the mainstream media) against previous years, usually by saying that month X has been the second (or third, take your pick) warmest of that month on record, often for a location of their choice that happens to generate a record this period. This can generate a succession of scary sounding headlines of rising temperatures comparing many different times and different locations. That in itself is cherry-picking data, by selecting the exact comparison to make after the results are all in and not before. For every scary temperature record report, there are a thousand that aren’t a record and therefore not interesting to report.
What I have NEVER seen is a story where any particular period exceeded the results predicted in a forecast. Think about that: has anyone ever seen an article in a journal or the mainstream media where actual temperatures anywhere in the world were higher than the climate models predicted? If that happened with my financial forecasting, I would be getting feedback (let the reader understand the euphemism) from senior management and head office that my forecasting was no good, but there appears to be no mechanism whereby climate scientists are similarly routinely called to account, so this essay is an attempt to do so exactly that.
The title of this essay
I have titled this essay as “The dog that never barks” as a mis-quotation from the Sherlock Holmes story Silver Blaze. In the story, Holmes solves the case because he, and only he, realises that the guard dog not barking during the night when a crime was committed meant that the guard dog knew the intruder and was not alarmed by him, so the lack of barking was the relevant fact. No-one else understood that the missing bark was the relevant fact.
In our present case, the lack of actual temperatures exceeding those forecast by climate models is the relevant fact. Not just once, as in the Sherlock Holmes story, but NEVER. The mainstream media, used to being fed press releases from climate activists rather than thinking for themselves, haven’t noticed the lack of comparison. This comparison against forecast is the best feedback there is. The implications of the findings when comparisons are made are staggering.
A Paper reporting comparisons
A very few comparisons are made and published in respectable journals, written by scientists truly interested in the reliability of climate models. These all find that the models overstate the rise in global temperatures that actually occur. For example, McKitrick & Christy1 examined a group of climate models in the Coupled Model Intercomparison Project Version 6 (CMIP6) and found that: “All model runs warmed faster than observations in the lower troposphere and mid-troposphere, in the tropics and globally. On average, and in most individual cases, the trend difference is significant. Warming trends in models tends to rise with the model Equilibrium Climate Sensitivity (ECS) and we present evidence that the distribution of ECS values across the model is unrealistically high.” That quote is taken from their Plain Language Summary but in even plainer language I understand that to mean the the models run too hot and the parameters chosen by the modellers are set too high.
Sometimes a picture is worth a thousand words. Figure 4 from McKittrick & Christy’s paper shows the predicted and actual warming per decade.
Note that there is no playing with axes, both ECS and warming have zero as the lowest value. The open shaped symbols concern the mid- troposphere, the closed ones the lower troposphere; red vs blue simply shows high and low ECS values used in the models. The really interesting points are the two inverted triangles at the foot of the graph where ECS is equal to zero, highlighted by the downward arrows pointing to them. These are the actual values observed by a variety of methodologies for the mid and lower troposphere. There is no overlap of warming trends found in actual values with the models, in either mid or lower troposphere. The vast majority of the data points for predicted temperature rise from the models far exceed what has been actually observed.
The even plainer statement of conclusions to draw from their paper is this: climate models have been shown to be highly inaccurate and to predict temperature rises well in excess of those found. No reliance should therefore be placed on their predictions of the future temperatures.
The Science is never settled
Climate campaigners like to avoid revisiting issues by saying “The science is settled”. But this viewpoint is very much non-scientific. The best synopsis of how scientists should behave was devised by sociologist Robert Merton around 1943, the so-called Mertonian Norms.2 This set of ideals takes the goals and methods of science that should be binding on scientists. His list comprises:
- Universalism: Scientific knowledge is scientific knowledge, no matter who comes up with it, as long as their methods are sound.
- Disinteredness: Scientists are not in it for the money, for political or ideological reasons or to enhance their own or their institution’s reputations. They should be doing science to advance our understanding of the universe.
- Communality: Scientists should share knowledge with each other. This is the reason that findings are published in journals.
- Organised Scepticism: Nothing is sacred, a scientific claim should never be accepted at face value. We should suspend judgement on any given finding until we have properly checked all the data and the methodology. The most obvious example of organised scepticism is the practice of peer review.
Ritchie shows that mainstream science has problems with the ideals of the Mertonian Norms, with many examples of fraud, bias, negligence, peer review problems, poor use of statistics (e.g. p-hacking) and perverse incentives in funding and hiring.
Climate science almost seems to delight in ignoring them or doing the exact opposite. Watts Up With That provides many, many examples of this.
The environmental campaign leading to demands for radical changes to our lifestyles relies on a progression from one assertion to another, all supposedly scientifically supported. This series starts with the assertion that climate models are a good predictor of future climate and temperature. The next assumption (probably the largest) is that carbon dioxide is the main, if not the only driver, of these climate models. The next assumption is that a rise in global temperatures in excess of 1.5 degrees Centigrade by 2050 will have a disastrous effect on the ecosystems of the planet. The final assumption in the chain is that limiting carbon dioxide generated by humanity to arbitrarily set limits will stabilise global temperatures.
My essay here shows that the first assumption is false. Climate models in the past have not predicted current temperature trends, so cannot be assumed to be accurate in the future. I could have chosen (and others probably will) to write and challenge any of the other assumptions in the chain, for all are equally erroneous.
The environmental movement is therefore inherently political, being based on unsound science, and should be open to robust challenge on both scientific and political grounds.
1. “Pervasive Warming Bias in CMIP6 Tropospheric Layers” Earth and Space Science 8, (7)
2. “Science Fictions, exposing Fraud, Bias, Negligence and Hype in Science” by Stuart Ritchie, Vintage, 2021 p.21